Data-driven approaches for proactive maintenance planning of sewer blockage management

Blockages have been reported to account for a significant proportion of reported failures in sewer networks. The malfunctioning of the sewer network from blockages and the subsequent disruption to other public services and flooding may constitute a risk to the environment and human health. Due to the complex nature of underground sewer networks, a reactive approach to blockage maintenance is typically employed. However, although proactive maintenance strategies have been developed, both approaches could be expensive and highlight the need to address the problem with analytics-based methods. Although blockage triggering mechanisms may be known, sewer blockages often appear at random. Thus, it is necessary to improve the understanding of the influential mechanisms involved in forming blockages in sewer networks to support its maintenance and guarantee adequate performance levels. The overall aim of this thesis was to contribute with new knowledge, approaches and methods that can support improved proactive maintenance planning of blockages in sewer networks. Various methods to achieve the aim have been investigated in relation to asset management planning levels. At the strategic level, blockages and associated performance indicators were employed in conjunction with Poisson and partial least squares regression to assess the performance of sewer networks, including gaining additional insights. At the tactical and operational levels, a procedure was developed. The procedure combines network k-function, geographically weighted regression and random forest ensembles. The network k-function analysis explains the significance of the spatial variation of blockages. The Geographically weighted Poisson regression (GWPR) investigates the degree of influence of explanatory factors on increased blockage propensity in differentiated segments of the sewer networks. Thirdly, the random forest ensembles was used to predict pipes with blockage recurrence likelihood. A proposed conceptual framework was applied at all asset management levels to assess the state of data-driven integrated asset management (IAM), based on data quality assessments, interoperability evaluations between IAM tools, and data collection and informational benefits analysis. Results from demonstrating the methods with data from the Swedish waters statistical database and three Swedish municipal sewer networks, namely A, B and C, are presented. Blockage related performance indicators showed that the average blockage rate in medium sized networks was 2-3 times the rate in other sewer networks in Sweden. Furthermore, sewer maintenance strategies were suspected to be ineffective, and increased proactive strategies may improve maintenance efficiency. The procedure in networks A, B and C indicated that the clustering of recurrent blockages maybe linked to an increased need for flushing-related maintenance in sewer pipe networks. The degree of influence between investigated factors and increased blockage propensity indicated that these relationships were not global (not the same in all locations) within and between the sewer networks for networks A, B and C. These non-stationary relationships were observed to occur in various forms, i.e. adequate self-cleaning velocity showed positive and negative correlations in different locations. The networks with relatively more substantial spatial clusters of blockages, higher data quality and availability were observed to have a higher mean prediction accuracy. The applied conceptual framework showed that intuitive asset management characterised the current state of blockage management in the municipal sewer network C with medium to good data quality and low interoperability